INDEX

Explanations

phrases indicating a critical or decisive moment

oai_token-act-pair · gpt-3.5-turbo

phrases indicating specific situations or conditions described with the word "where."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Bees

-0.70

rolet

-0.69

yi

-0.67

 Deer

-0.62

fml

-0.61

apult

-0.61

 Beasts

-0.61

mask

-0.60

cade

-0.60

Bow

-0.60

POSITIVE LOGITS

upon

1.35

abouts

1.05

soever

0.96

nces

0.71

ILLE

0.71

imaru

0.69

fore

0.69

hov

0.68

earch

0.68

reon

0.67

Activations Density 0.052%

phrases indicating a critical or decisive moment

phrases indicating specific situations or conditions described with the word "where."

No Comments

No Known Activations